Download Melting features along the Ryukyu slab tear, beneath the

GEOPHYSICAL RESEARCH LETTERS, VOL. 31, L19607, doi:10.1029/2004GL020862, 2004
Melting features along the Ryukyu slab tear, beneath the southwestern
Okinawa Trough
Jing-Yi Lin,1 Shu-Kun Hsu,2 and Jean-Claude Sibuet1
Received 28 June 2004; revised 20 August 2004; accepted 16 September 2004; published 8 October 2004.
[1] The present-day active volcanic front associated with
the Ryukyu subduction zone extends from Japan to the Ilan
plain (northern Taiwan) and is located within the Okinawa
Trough, 80– 100 km above the Ryukyu slab. An abnormal
amount of arc volcanism, which consists of basalt, andesite
and rhyolite occurs within the southwestern Okinawa
Trough, above a slab tear of the Ryukyu subduction zone
(CBVT). The power spectrum analysis of magnetic data
shows the occurrence of a thin crust above the slab tear and
a thick crust beneath this volcanic area. We suggest that an
excess of H2O-rich fluid might occur at the slab tear and
might increase the melt flux. Both are conveyed obliquely
to the uppermost mantle and lower crust CBVT magmas.
After interactions, basaltic magmas would rise up,
accounting for the contrast of magnetization between this
INDEX TERMS:
volcanic body and the adjacent OT crust.
1517 Geomagnetism and Paleomagnetism: Magnetic anomaly
modeling; 7218 Seismology: Lithosphere and upper mantle; 8100
Tectonophysics; 8159 Tectonophysics: Rheology—crust and
lithosphere. Citation: Lin, J.-Y., S.-K. Hsu, and J.-C. Sibuet
(2004), Melting features along the Ryukyu slab tear, beneath the
southwestern Okinawa Trough, Geophys. Res. Lett., 31,
L19607, doi:10.1029/2004GL020862.
1. Geological and Petrological Settings
[2] The Ryukyu subduction zone, which extends from
Japan to Taiwan, terminates westwards beneath northeastern
Taiwan (Figure 1). The volcanic front (dotted gray line in
the upper inset of Figure 1) also extends from Japan to
Taiwan and is located within the Okinawa Trough (OT).
From Kyushu to Okinawa Island, it coincides with small
subaerial active volcanoes located about 25 km west of the
non-volcanic arc. Southwest of Okinawa Island, it follows
numerous seamounts associated with high magnitude magnetic anomalies [Hsu et al., 2001], then the cross backarc
volcanic trail (CBVT) which consists of a cluster of about
70 seamounts located west of 123E longitude (lower inset
in Figure 1) and finally ends at Kueishantao Island [Chung
et al., 2000; Sibuet et al., 1998], an islet located 10-km
offshore the Ilan Plain (OT tip). The volcanic front is 80–
100 km above the subducted slab. The minimum slab depth
of 80 km is required for the emplacement of arc magmatism
[Gill, 1981; Tatsumi, 1986], explaining why the present-day
Ryukyu arc is not located within the Ryukyu non-volcanic
islands but within the OT.
1
Institut Français de Recherche pour l’Exploitation de la Mer, Centre de
Brest, Plouzané, France.
2
Institute of Geophysics, National Central University, Chung-Li,
Taiwan.
Copyright 2004 by the American Geophysical Union.
0094-8276/04/2004GL020862$05.00
[3] Amongst the several tens of volcanic seamounts
identified in the CBVT [Sibuet et al., 1998], most of
them are located in the vicinity of the axial depression.
Basalts, andesites and rhyolites collected there belong to
the medium-K field in the SiO2vs K2O diagram and are
subalkaline [Chung et al., 2000; Shinjo et al., 2003a, 2003b].
Rhyolites are dominant in the CBVT. Shinjo et al. [2003a,
2003b] demonstrate that mafic intrusions in the CBVT
rhyolites resulted from magma mixing between high temperature (high-Mg content) mafic and low temperature (low-Mg
content) felsic magmas, implying a high degree of partial
melting with slab components. The incompatible elements
are enriched in large ion lithophile elements, Th, U and Pb
and depleted in Nb, Ta and Ti [Chung et al., 2000; Shinjo et
al., 2003a, 2003b]. Thus, the elemental and isotopic variations are compatible with magma derivation from an Indian
NMORB-like source that is strongly overprinted by subduction components from the slab [Shinjo et al., 2003a, 2003b].
Thus, CBVT basalts are not typical of backarc basin basalts.
2. Ryukyu Slab Tear and Seismicity
[4] Relocated earthquake hypocenters from teleseismic
data recorded from 1964 to 1999 [Engdahl et al., 1998] with
a standard deviation less than 20 km are plotted for 50-km
wide bandwidths on each side of the northern prolongation of
the Huatung Basin/west Philippine (PH) Sea shear plate
boundary located east of Gagua Ridge and underlined by a
negative free-air anomaly trend [Hsu et al., 1998] (Figures 1
and 2c). The upper envelope of these earthquakes corresponds to the top of the subducted slab. As already mentioned
by Deschamps et al. [2000], the portion of slab located north
of the Huatung Basin (A) presents a steeper dipping angle
than the one located north of the west Philippine Basin (B).
However, we suggest that the PH plate was torn rather than
bent during the subduction process for the following reasons:
(1) The Huatung Basin/west (PH) Sea plate boundary is a
zone of weakness [Sibuet et al., 2002]. (2) The lithospheres
on each side of this boundary are characterized by different
crustal thicknesses and rheologies [Hsu, 2001; Murauchi et
al., 1968; Wang et al., 2002]. (3) At 25.5N latitude, the
western portion A of the Ryukyu slab is 80 km deeper than
the eastern portion B of the slab (Figures 1 and 2c). (4) The
slab displays different strain patterns on each side of 123.3E
[Kao and Chen, 1991]. We consequently suggest a link
between the location of the Ryukyu slab tear and the localized
excess of volcanism in the CBVT region, whose geochemical
composition reflects a slab component and arc signature.
3. Depth and Thickness of the Magnetized Crust
[5] The southern Okinawa Trough exhibits low magnetization values [Hsu et al., 2001] except for a relative higher
L19607
1 of 4
L19607
LIN ET AL.: MELTING FEATURES ALONG THE RYUKYU SLAB TEAR
Figure 1. Bathymetry, topography (every kilometer with
shading from N315) and seismicity [Engdahl et al., 1998]
in the northwestern corner of the Philippine Sea plate. Dots
represent earthquake hypocenters. The thick contour lines
are depths of the Ryukyu slab (adapted from Font et al.
[1999] in the western part). The dashed line shows the
location of the Huatung Basin/western PH Sea plate
boundary underlined by a negative free-air anomaly trend
[Hsu et al., 1998] and located just east of Gagua Ridge. Its
northern prolongation coincides with the Ryukyu slab tear,
underlined by the offsets of slab isobaths. In the upper left
corner, general map of the Ryukyu subduction zone with
isobaths of the Ryukyu slab every 50 km [Sibuet et al.,
1998]. The volcanic front is located 80– 100 km above the
slab. In the lower right corner, detailed bathymetry (isobath
spacing, 100 m) of the cross backarc volcanic trail (CBVT)
[Sibuet et al., 1998] in the square box (dashed black line)
located west of the slab tear. The arrow indicates the PH/EU
plate motion from Yu et al. [1997].
magnetization zone located between 24.9N and 25.5N
[Hsu et al., 2001] (Figure 2b), west of the slab tear. To
understand the thickness variation of the magnetic source
beneath the southern Okinawa Trough, we use the power
spectrum technique to estimate the depths to the top (Zt) and
to the centroid (Z0) of the magnetic source. If the wave
number (k) is less than about twice the thickness of the
magnetized layer, Zt is calculated from the slope between
Ln[FDT(jkj1/2)] and jkj, where the FDT(jkj) is the powerdensity of the averaged radial spectrum of the total-field
magnetic anomaly. At longer wavelengths, Z0 is given by
the slope between Ln{[FDT(jkj)1/2]/jkj} and jkj [Blakely,
L19607
Figure 2. a) Magnetic anomalies from Hsu et al. [2001].
Thin black lines are bathymetric contours every kilometer.
The square box in dashed black lines (CBVT, cross backarc
volcanic trail) corresponds to the location of the bathymetric
inset in Figure 1. b) The portion of the Okinawa Trough
deeper than 1000 m is characterized by low magnetization
values except for the higher magnetization zone located
west of the Ryukyu slab tear (white dashed line).
c) Earthquake distribution within two 50-km wide parallel
stripes located on each side of the 123.3E meridian.
Crosses and open circles represent earthquakes located west
and east of this limit respectively. The western portion A of
the slab is steeper and deeper than the eastern portion B of
the slab. d) Estimates of the depth and thickness of magnetic
sources (thick lines) extracted from Figures e and f.
Comparison with the P-wave velocity-interface model along
refraction Profile 1 [Wang et al., 2002] (thin dotted/dashed
lines). e) Depth Zt of the top of magnetic sources using the
power spectrum method. Values are calculated every
0.2 degree in 60 km 60 km data squares; f) Depth Zb
of the base (Curie point) of the magnetic sources using the
same method; g) thickness of the magnetized layer (Zb-Zt).
h) Geometry of the crust deduced from Zt and Zb
distributions along Profile 2.
2 of 4
L19607
LIN ET AL.: MELTING FEATURES ALONG THE RYUKYU SLAB TEAR
1988; Tanaka et al., 1999]. The basal depth Zb of the
magnetic source, assumed to be the Curie point depth, is
then calculated by Zb = 2Z0 Zt.
[6] In order to enhance the broad features linked to deep
features, the centroid depths were calculated for wavelengths larger than 10 km [Stampolidis and Tsokas, 2002;
Tanaka et al., 1999]. Zt, Zb and the thickness of the
magnetized crust (Zb-Zt) have been computed on a regular
grid pattern with a 0.2 degree spacing. The size of each
calculated area is 60 km 60 km. For reasonable centroid
estimates, a minimum ratio of 12 between the size of the
used data square and the depth of the centroid point was
established by Okubo et al. [1985]. Figure 2d shows a
comparison between the estimated depths Zt and Zb with the
P-wave velocity model computed from the refracted and
reflected arrivals along Profile 1 [Wang et al., 2002]. Zt and
Zb correspond approximately to the 4.5 (top of volcanics)
and 7.75 (Moho) km/s interfaces, which gives credit to the
maps of Figures 2e – 2g.
[7] The map of magnetic anomaly data (Figure 2a) shows
the presence of a large magnetic body located in the axial
part of the OT, west of 123.3E and at an average depth of
3 – 4 km (Figure 2h). The CBVT seamounts, identified at
mean depth of 1.3 km, are shallower than this estimate. This
difference might be explained by the smoothing effect due
to the power spectrum estimation. However, the seismic
reflection [Sibuet et al., 1998] and diving [Matsumoto,
2001] data show that the CBVT seamounts have been
intruded through a 1-km thick sedimentary cover and that
a significant amount of non-magnetized hydrothermal
deposits exists on top of some of them. At depth, the Curie
point surface (Figure 2h), which is close to the Moho
surface, is about 25 km beneath the CBVT and 15 km at
the vertical of the slab tear, above which no volcanism was
evidenced (Figure 2g). Thus, the crust is thick beneath the
CBVT and thin beneath the slab tear (Figures 2g and 2f).
4. Discussion and Conclusion
[8] In the southwestern OT, only characterized by a few
thin elongated backarc basin features, the CBVT appears as
an abnormal area of Recent and Present-day volcanism
[Sibuet et al., 1998]. In general, most arc volcanic rocks
are derived from melting of the mantle wedge induced by
hydrous fluids released during dehydration reactions in the
subducted oceanic lithosphere [Arculus, 1994; Gill, 1981].
Melt and/or fluid flow from the underlying slab through an
oblique mantle wedge and crustal pathway have been
spectacularly imaged by tomographic data both beneath
the Quaternary arc volcanoes of Japan [Wyss et al., 2001]
and beneath Kueishantao Island (J. Y. Lin et al., Melting
features along the western Ryukyu slab edge (northeast
Taiwan): Tomographic evidence, submitted to Journal of
Geophysical Research, 2004, hereinafter referred to as Lin
et al., submitted manuscript, 2004). The CBVT melt or fluid
supply could have a westward oblique component originated
from the slab tear, even if a significant part of the melt was
formed within the upper mantle and lower crust, as a
consequence of the OT backarc basin extension.
[9] We suggest that the deep process, which is proposed
for the origin of Kueishantao Island located close to the
western termination of the Ryukyu slab (Lin et al., submitted
L19607
Figure 3. Diagram showing the Ryukyu slab tear
occurring along the northern prolongation of Gagua Ridge,
in the northwestern corner of the Philippine Sea plate.
Stippled arrows indicate fluid migrating pathways around
the slab edges. Contorted stippled arrows give an indication
of the oblique fluid and melt pathway to the cross backarc
volcanic trail (CBVT) and Kueishantao Island. Black arrow
shows PH/EU plate motion.
manuscript, 2004), could be applied for the CBVT, located
140 km east of the Ryukyu slab termination (Figure 3). The
eastern part B of the Ryukyu slab being shallower than the
western part A, two hypotheses might be proposed:
(1) Abnormal heating due to the friction of the vertical
portion of slab B (located above the portion of slab
A) against the adjacent EU lithosphere and/or the upwelling
of the underlying PH mantle material through the slab tear
could induce the formation of additional melt above the
slab B. However, Rüpke et al. [2004] suggest that shear
stresses and shear heating, though largely unconstrained, are
weak at the boundary between the slab and the overlying
mantle and lithosphere. Thus, friction heating might be
negligible. Upwelling of underlying PH mantle might only
occur north of 25.5N where the PH mantle of slab B largely
culminates above slab A. Such a mantle upwelling would
occur about 50 km north of the CBVT, which might require a
too much oblique feeding pathway to reach the CBVT
magmas. (2) Increased H2O-rich component derived from
fluids from the underlying subducting slab and slab tear
might enhance the generation of CBVT magmas located in
the EU upper mantle/lower crust [Stolper and Newman,
1994]. Backarc basin magmas worldwide are rich in H2O
relative to NMORB, so a general mechanism must exist (as
an upward migration of fluids generated by deep dehydration) to account for the water enrichment. Fluid release
occurs above the slabs at depths <20 km from subducting
sediments, at intermediate depths (20 – 100 km) from sediments and oceanic crust and at depths >100 km from oceanic
crust and serpentinized mantle [Rüpke et al., 2004]. An
3 of 4
L19607
LIN ET AL.: MELTING FEATURES ALONG THE RYUKYU SLAB TEAR
additional amount of fluid might be released in the slab tear
region, along the vertical portion of slab B located above
slab A. In that case, the H2O-rich fluid would be conveyed
from around the edge of slab B in direction of the CBVT
magmas. After fluid-mantle interactions and introduction of
fluids into basaltic magmas, mostly produced in the uppermost mantle and lower crust, magmas would rise up along
veins and conduits into the overlying upper crust of the
CBVT area, accounting for the contrast of magnetization
between this volcanic body and the adjacent OT crust.
However, it is widely accepted that an increase of water
amount in the mantle wedge results in an increase of the
degree of melting and melt flux [Stolper and Newman,
1994]. A combination of higher water flux and melt flow
in the mantle wedge, possibly driven by the geometry of the
slab tear, might account for the abnormal magmatism in the
CBVT region. Such a model is illustrated in the sketch of
Figure 3.
[10] Acknowledgments. The GMT software package was used to
make some of the figures [Wessel and Smith, 1991]. This work was carried
out within the frame of the Sino-French cooperation in Geosciences. We
thank the National Science Council (NSC) of Taiwan and the Institut
Français à Taipei (IFT) for their financial support during the Ph. D. thesis of
one of us (J.-Y. Lin). Pertinent discussions with S.-L. Chung, L. Ciron,
L. Dosso, C.-S. Lee and W. Roest are acknowledged as well as a very
constructive and helpful review from R. Shinjo.
References
Arculus, R. J. (1994), Aspects of magma genesis in arcs, Lithos, 33, 189 –
208.
Blakely, R.-J. (1988), Curie temperature isotherm analysis and tectonic
implications of aeromagnetic data from Nevada, J. Geophys. Res., 93,
11,817 – 11,832.
Chung, S.-L., S.-L. Wang, R. Shinjo, C.-S. Lee, and C.-H. Chen (2000),
Initiation of arc magmatism in an embryonic continental rifting zone of
the southermost part of Okinawa Trough, Terra Nova, 12, 225 – 230.
Deschamps, A., P. Monié, S. E. Lallemand, S.-K. Hsu, and K. Y. Yeh
(2000), Evidence for early Cretaceous oceanic crust trapped in the
Philippine Sea plate, Earth Planet. Sci. Lett., 179, 503 – 516.
Engdahl, E. R., R. D. Van der Hilst, and R. P. Buland (1998), Global
teleseismic earhquake relocation with improved travel times and procedures for depth determination, Bull. Seismol. Soc. Am., 88, 722 – 743.
Font, Y., S. Lallemand, and J. Angelier (1999), Etude de la transition entre
l’orogène actif de Taiwan et la subduction des Ryukyu—Apport de la
sismicité, Bull. Soc. Géol. France, 170, 271 – 283.
Gill, J. (1981), Orogenic Andesite and Plate Tectonics, 390 pp., SpringerVerlag, New York.
Hsu, S. K. (2001), Lithospheric structure, buoyancy and coupling across the
southermost Ryukyu subduction zone: An example of decreasing plate
coupling, Earth Planet. Sci. Lett., 186, 471 – 478.
Hsu, S.-K., C.-S. Liu, C.-T. Shyu, S.-Y. Liu, J.-C. Sibuet, S. Lallemand,
C. Wang, and D. Reed (1998), New gravity and magnetic anomaly maps
in the Taiwan-Luzon region and their preliminary interpretation, Terr.
Atmos. Ocean Sci., 9, 509 – 532.
L19607
Hsu, S.-K., J.-C. Sibuet, and T.-C. Shyu (2001), Magnetic inversion of
the East China Sea and Okinawa Trough: Tectonic implications,
Tectonophysics, 333, 111 – 122.
Kao, H., and W.-P. Chen (1991), Earthquakes along the Ryukyu-Kyushu
arc: Strain segmentation, lateral compression, and thermomechanical
state of the plate interface, J. Geophys. Res., 96, 21,443 – 21,485.
Matsumoto, T. (2001), Volcanic and hydrothermal activities and possible
‘‘segmentation’’ of the axial rifting in the westernmost part of the
Okinawa Trough—Preliminary results from the Yokosuka/Shinkai 6500
Lequios cruise, JAMSTEC J. Deep Sea Res., 19, 96 – 107.
Murauchi, D., et al. (1968), Crustal structure of the Philippine Sea,
J. Geophys. Res., 73, 3143 – 3171.
Okubo, Y., R.-J. Graf, R.-O. Hansen, K. Ogawa, and H. Tsu (1985), Curie
point depths of the island of Kyushu and surrounding areas, Japan, Geophysics, 50, 481 – 494.
Rüpke, L. H., J. Phipps Morgan, M. Hort, and J. A. D. Conolly (2004),
Serpentine and the subduction zone water cycle, Earth Planet. Sci. Lett.,
223, 17 – 34.
Shinjo, R., S. Hokakubo, S. Haraguchi, and T. Matsumoto (2003a), Regional
variation in geochemistry of volcanic rocks from the southern Okinawa
Trough (in Japanese), Earth Mon., 43, 21 – 26.
Shinjo, R., S. Hokakubo, S. Haraguchi, T. Matsumoto, and J. Woodhead
(2003b), Geochemical characteristics of volcanic rocks from the southern
Okinawa Trough and its implications for tectono-magmatic evolution,
Eos Trans. AGU, 84(46), Fall Meet. Suppl., Abstract V31E-0973.
Sibuet, J.-C., and S.-K. Hsu (2004), How was Taiwan created?, Tectonophysics, 379, 159 – 181.
Sibuet, J.-C., B. Deffontaines, S.-K. Hsu, N. Thareau, J.-P. Le Formal, and
C.-S. Liu (1998), The southwestern Okinawa Trough backarc basin:
Tectonics and volcanism, J. Geophys. Res., 103, 30,245 – 30,267.
Sibuet, J.-C., S.-K. Hsu, X. Le Pichon, J.-P. Le Formal, D. Reed, G. Moore,
and C.-S. Liu (2002), East Asia plate tectonics since 15 Ma: Constraints
from the Taiwan region, Tectonophysics, 344, 103 – 134.
Stampolidis, A., and G.-N. Tsokas (2002), Curie point depths of Macedonia
and Thrace, N. Greece, Pure Appli. Geophys., 159, 2659 – 2671.
Stolper, E., and S. Newman (1994), The role of water in the petrogenesis of
Mariana trough magmas, Earth Planet. Sci. Lett., 121, 293 – 325.
Tanaka, A., Y. Okubo, and O. Matsubayashi (1999), Curie point depth
based on spectrum analysis of the magnetic anomaly data in East and
Southeast Asia, Tectonophysics, 306, 461 – 470.
Tatsumi, Y. (1986), Formation of the volcanic front in subduction zones,
Geophys. Res. Lett., 13, 717 – 720.
Wang, T. K., K. McIntosh, Y. Nakamura, C.-S. Liu, and H.-W. Chen
(2002), Velocity-interface structure of the southwestern Ryukyu subduction zone from EW9509 – 1 OBS/MCS data, Mar. Geophys. Res., 22,
265 – 287.
Wessel, P., and W. M. F. Smith (1991), Free software helps map and display
data, Eos Trans. AGU, 72, 441 – 446.
Wyss, M., A. Hasegawa, and J. Nakajima (2001), Source and path of
magma for volcanoes in the subduction zone of northeastern Japan, Geophys. Res. Lett., 28, 1819 – 1822.
Yu, S.-B., H. Y. Chen, and L.-C. Kuo (1997), Velocity field of GPS stations
in the Taiwan area, Tectonophysics, 274, 41 – 59.
S.-K. Hsu, Institute of Geophysics, National Central University, Chung-Li
32054, Taiwan. ([email protected])
J.-Y. Lin and J.-C. Sibuet, Ifremer, Centre de Brest, B.P. 70, 29280
Plouzané Cedex, France. ([email protected]; [email protected])
4 of 4